To Budget or Not to Budget – The AI Perspective
Read-on if your marketing budget decisions give an impression of being based on Roulette wheel outcome. Read-on if your monthly marketing budgets are “locked” by channels. Read-on if you think AI is super cool and will help you deliver ROI against your spends. AI skeptics Read-on!
Budgeting Decisions can be extremely complex when we consider multiple moving parts of the marketing engine – channels/sub-channels, Ad types, campaigns types, categories of products, transaction volumes, attribution models.
Faced with this complexity, it is common today to have marketing budgets finalized at the start of each month. No real surprise that these digital marketing strategy sessions reflect the age-old “media planning” sessions.
AI, driven by intelligent machine learning algorithms, can fill in the gap by delivering a highly agile and rewarding experience for your marketing team. Successful machine learning algorithms rely on capturing and analysing data on multiple axis.
Continuing with the budgeting decision making example, platforms will deploy basic optimization techniques like Hill Climb and Random Search algorithms when the dataset is fairly linear. A typical scenario will be for an algorithm deciding on the best performing ad copies based on the combination of ad placement, images, titles, descriptions and landing pages.
Hill climbs have an inherent weakness where the algorithm breaks on a local maxima. Any smart platform will minimize the impact by implementing the random restart technique.
As the complexity increases e.g. when we start adding multiple channels, campaigns, attribution models in the mix, platforms switch to more advanced optimization techniques including Annealing and Genetic algorithm.
Where Hill climb relies on the one-to-one comparison technique, Annealing allows for probability of success even for a relatively lower scoring combination. To elaborate, let’s consider we have 50 different campaigns with 5 ad sets each. There is a success score assigned to each moving component of the campaign and ad set – audience age, gender, education, interests, website behaviour etc.
With the Hill climb algorithm the individual component scores will be added and the campaign budgets will be decided by basic arm-wrestling elimination technique.
Here the “losing” campaign has no chance of getting a larger budget. With Annealing though, the loser still has a chance for a success or a higher budget as long as it can be within the winning probability
P = exp(-c/t) > r
c = the change in the evaluation function
t = the current temperature
r = a random number between 0 and 1
The success of AI platforms is based on their ability to self-learn and improve on any of the assumptions and variables. With ever increasing computational ability, AI algorithms can drive real-time budgeting.
Budgets are at the core of any marketing success, and it will be useful for you to take a real-time decision to invest in an AI-Driven E-Commerce Marketing platform. You still reading? Go get one! 🙂